Hi Jie-Yu, Yes, that looks fine if this design will be used with FEAT. All the best, Anderson On 28 February 2015 at 13:20, Jie Yu <[log in to unmask]> wrote: > Dear Expert > > So I would like to know the mean activation among 2 groups (with different > variances within groups). > > Also, I will have age as a covariate. > > > My guess is this: > > * The first 2 inputs are from group A and the last 2 inputs are from group > B > > * EV1,EV2 are groups whereas EV3,EV4 are ages. > > Group EV1 EV2 EV3 EV4 > 1 1 0 23 0 > 1 1 0 22 0 > 2 0 1 0 30 > 2 0 1 0 23 > > Contrast1 0.5 0.5 0 0 (mean activation) > Contrast2 -0.5 -0.5 0 0 (mean deactivation) > > Hope that's ok. > > Thanks so much! >